Antagonism of Trichoderma on the control of Fusarium spp. on Phaseolus lunatus L

Biological control strategies have become an important tool in the sustainable management of plant diseases. This paper aims to report the Fusarium species that affect fava beans (Phaseolus lunatus L.) grown in Paraíba, Brazil, and determines the potential of Trichoderma isolates to control these fungi. Two Trichoderma and ten Fusarium isolates from fava bean seeds were selected. The beans were obtained from cultivated areas in the municipalities of Remígio, Alagoa Grande and Campina Grande, in Paraíba state. Phylogenetic analyzes based on DNA sequences of the translation elongation factor 1-α (TEF1) gene resolved the Fusarium isolates into four species belonging to the F. fujikuroi and F. incarnatum-equiseti species complexes. In vitro tests showed that the two isolates of Trichoderma tested presented antagonistic potential against the pathogens from the fava beans evaluated. In the direct comparison test, the growth of the pathogens was reduced from the seventh day in both treatments. Sporulation also showed a reduction, but only for 40% of Fusarium isolates. This work demonstrates that Trichoderma isolates can be used as a sustainable alternative to manage Fusarium spp. infection of fava beans.


Introduction
Fava bean seeds (Phaseolus lunatus L.) are grown worldwide, due to their high protein and nutrient content. It is one of the species of greatest economic and social importance for the Brazilian Northeast (Santos, 2008). The Paraíba state accounts for almost 50% of the production of this legume in the Northeast region and is cultivated mainly by family farmers (Gomes, Nunes, Nascimento, Souza & Porcino, 2016). However, fava bean productivity has been reduced by diseases (Gomes & Nascimento, 2018). The seeds can carry pathogens and introduce them into new production areas (Silva-Flávio, Sales, Aquino, Soares, Aquino & Catão, 2014;Nascimento & Medeiros, 2015).
The Fusarium genus is an important plant pathogen that infects many economically important crops (Summerell, 2019). Vascular wilts, cankers, stem, and root rots are some of the symptoms caused by Fusarium in plants (Summerell & Leslie, 2011). Fusarium currently comprises hundreds of species distributed in 23 species complexes (Summerell, 2019). Many of these species are morphologically cryptic and require molecular methods for their identification (O'Donnell et al., 2015). The inability to distinguish cryptic species among pathogens complicates disease management (Bickford, Lohman, Sodhi, NG, Meier, Winker, Ingram & Indraneil, 2007). Thus, accurate identification of Fusarium species is essential for control strategies to be implemented properly (Santos, Trindade, Lima, Barbosa, Costa, Carneiro-Leão & Tiago, 2019).
Biological control is used as an effective alternative for plant pathogens control. Microorganisms used in biological control exhibit antagonistic characteristics and can increase resistance against pathogens (Bhattacharryya, Goswami & Bhattacharyya, 2016). Trichoderma species are promising to plant pathogen management strategies because they are ecofriendly and sustainable (Mousumi Das, Haridas & Sabu, 2019), avirulent plant symbionts, and common on rhyzosphere. They are known by their antagonism against phytophatogens and mycotoxin production (Harman, Howell, Viterbo, Chet & Lorito, 2004;Zhang et al., 2017). There are many reported compounds associated with Trichoderma, such as acetylorcinol, alternariol, cerevisterol and scytalones, that causes inhibitory effect on other microrganisms (Zhang et al., 2017). Nowadays, more than 60% of biopesticides contain Trichoderma isolates in their composition, which put them among the most explored biocontrol agents (López-Bucio, Pelagio-Flores & Herrera-Estrella, 2015).
The aim of this work was to identify and report species of Fusarium obtained from fava beans seeds in the state of Paraíba, Brazil, and to evaluate the in vitro potential of Trichoderma as a biological control agent against Fusarium

Origin of fava bean seeds
Creole fava beans seeds, variety "Orelha de Vó", were obtained from Alagoa Grande, Remígio, and Campina Grande counties, Paraíba, Brazil, from local producers and commercial markets.

Isolation and identification of Fusarium isolates
The health seeds were evaluated by the Blotter-test method using 10 Petri dishes (15 cm) previously sterilized, with filter paper moistened in distilled and sterilized water, containing 20 seeds per plate, totaling 200 seeds per lot. After seven days of incubation at 25 ± 2 ºC in a Biochemical Oxygen Demand (BOD) incubator, the somatic and reproductive structures of Fusarium were identified by optical and stereomicroscope and isolated on PDA medium. Ten Fusarium isolates used in this research, from common beans seeds (Phaseolus vulgaris L.), and described as Isolate 01 through Isolate 10.
For molecular identification of Fusarium isolates, the DNA was extracted using the Wizard Genomic DNA Purification Kit (Promega). Partial sequences of the chain elongation factor 1 alpha (TEF1) were amplified using primers EF1α and EF2 (O'Donnell et al., 1998) and the cycling conditions were 8 min at 95 C, followed by 35 cycles of 15s at 95 C, 20 s at 53 C, and 1 min at 72 C, and a final step for 5 min at 72 as described on Santos et al. (2019). The sequences generated were deposited in Genbank under accession numbers MW846625, MW846627, MW846631, MW846632, MW846628, MW846624, MW846633, MW846629, MW846626, MW846630, for the isolates 01 to 10, respectively. These sequences were used to perform BLASTn searches in NCBI's GenBank database to determine the complexes to which they belong. After that, sequences were obtained from the databases and aligned using MAFFT v. 7 (Katoh,, Rozewicki & Yamada, 2019). Phylogenetic reconstructions of maximum likelihood were performed using RAxML-HPC2 on XSEDE (8.2.10) (Stamatakis, 2014) on the portal CIPRES Science Gateway (Miller, Pfeiffer & Schwartz, 2012), with 1000 replications of bootstrap and GTR nucleotide replacement model.

Obtaining Trichoderma isolates
Two Trichoderma isolates were obtained from the Laboratory of Phytopathology of the Federal University of Paraíba: Trichoderma sp., isolated from cotton seeds (Gossypium sp.), and an isolate from the commercial product Trichodel ® (Trichoderma asperellum).

Evaluation of Trichoderma antagonism in vitro
The antagonism of Trichoderma against ten Fusarium isolates was evaluated in vitro by a direct confrontation assays in Petri dishes, according to Carvalho, Mello, Júnior & Silva. (2011). Discs (5mm) containing the pathogen, were transferred to Petri dishes (90 mm) with PDA medium, three days before antagonist incubation, in opposite positions on the plate. The plates were incubated at 25 ± 2 ºC and a 12hour photoperiod with fluorescent light.
The colony diameters were measured daily and, based on the index by Bell, Wells & Markham (1982), scores ranging from 1 to 5 were attributed: 1. Antagonist grows and occupies the entire plate; 2. Antagonist grows and occupies part of the pathogen, around 2/3 of the Petri dish; 3. Antagonist and pathogen grow to half of the plate where neither dominates the other; 4. The pathogen grows and occupies part of the antagonist, around 2/3 of the Petri dish; 5. The pathogen grows and occupies the entire Petri dish.
The mycelial growth rate index was evaluated according to Oliveira (1991). The Fusarium spore production was evaluated after the 13th day of incubation. The spores were released with ADE and Tween 20, with subsequent filtering in sterile gauze and adjusting the conidia concentration thru 1.0 x 10 4 conidia / mL -1 in the hemacitometer (Alfenas & Mafia, 2016).
The experimental design was completely randomized, with four replicates of both Fusarium and Trichoderma isolates (10x4). Four Petri dishes for each Fusarium isolate were used as a control, with 120 plates in total.

Results e Discussion
Ten Fusarium isolates were obtained from fava beans seeds and purified. In BLASTn searches on GenBank, using the generated TEF1 sequences, three isolates showed greater similarity with species sequences of the F. incarnatum-equiseti species complex (FIESC), while the other seven isolates showed greater similarity with sequences of the F. fujikuroi species complex (FFSC). Phylogenies were prepared for each one of these complexes.
In the FIESC, the phylogenetic analysis ( Figure 1) showed two isolates in the F. sulawesiensis species (= FIESC 16) and one isolate formed a genealogically exclusive lineage that may represent a new phylogenetically species belonging to F. incarnatum clade of the FIESC, however, further studies are necessary to present a proper morphological description and species delimitation. The FIESC comprises 38 species, and a few phylogenetically distinct strains (Xia et al., 2019), isolated from various biological sources, and often from soil and plants. Among these species are found mycotoxins producers that can cause food contamination and impact the humans and animals health (Avila et al., 2019).  In Brazil, F. sulawesiensis has been reported associated with rice (Oryza sativa L.) (Avila et al., 2019) and melon (Cucumis melo L.) (Lima et al., 2020). This paper represents the first record of F. sulawesiensis associated with fava beans. Other species of the FIESC complex have been recorded in other countries associated with Fabaceae species, for example, F. equiseti in Vicia faba seeds in Poland (Sadowski, 1988) and F. duofalcatisporum in P. vulgaris seeds in Sudan (O'Donnell et al., 2009).
In the FFSC, the phylogenetic analysis supported four isolates in a clade with sequences of specimens of F. verticillioides, a member of the African subclade of the FFSC; and three isolates as a possible new phylogenetic species, genealogically related to F. proliferatum and F. annulatum, Asian subclade of the FFSC members ( Figure 2). Further studies are necessary to present a proper morphological description and species delimitation. Fungi of this complex are known to causes diseases in grains, such as ear and stalk rot of maize (Zea mays L.), bakanae disease of rice (Oryza sativa L.), and deterioration of other crops, such as barley (Hordeum vulgare L.), soybean (Glycine max L.), in addition to producing mycotoxins (Choi et al., 2018).
Fusarium verticillioides is known as one of the main pathogens of maize plants, reducing grain productivity worldwide (Leslie & Summerell, 2006;Watson, Burgess, Summerell & O'keeffe, 2014). Mota et al. (2017) and Sousa et al. (2020) reported that F. verticillioides infects fava beans in Brazil, however, the diagnosis was made based only on morphological characteristics. Thus, this paper comprises the first record of F. verticillioides in fava beans in Brazil, based on phylogenetic analyzes.

Trichoderma antagonism analysis
In the direct confrontation test, the two Trichoderma isolates inhibited the total growth of Fusarium isolates, from the fifth or sixth day of incubation. On the 13th day, the two antagonists occupied the plate completely, in contrast to the most of Fusarium isolates (Table 1). Similar results were observed by Rodrigues, Magalhães, Costa & Luz (2018) where Trichoderma isolates grew faster than the pathogen, stabilizing their growth in a few days.
All Fusarium isolates were inhibited by treatments with Trichoderma compared to the control, both on the 7th and 13th day. Comparing the effect of Trichoderma sp. and T. asperellum on Fusarium isolates, on the 7th day, for isolates 04 (Fusarium sp. of the FFSC), 06 (Fusarium sp. of the FIESC) and 10 (F. verticillioides) the greatest inhibition was obtained using T. asperellum, and for isolates 01 (F. sulawesiensis) and 07 (Fusarium sp. of the FFSC), it was using Trichoderma sp. For the other five isolates, there was no statistically significant difference between the two treatments with Trichoderma. On the 13th day, there was a significant difference among the two treatments just for isolates 01 (F. sulawesiensis), 03 and 06 (Fusarium sp. of FIESC), which showed the highest inhibition when treated with Trichoderma sp (Table 1).
Comparing Fusarium isolates, on the seventh day, isolates 08 and 10 of F. verticillioides were highly inhibited by Trichoderma sp.; and isolate 08 by T. asperellum. On the 13th day, all Fusarium isolates responded to treatment with Trichoderma sp. similarly, but when Fusarium isolates were treated with T. asperellum differences were observed with the highest inhibitions showing on for all isolates of F. sulawesiensis from FIESC and, F. verticillioides, and for two of the three isolates of Fusarium sp. of the FFSC (Table 1). Averages followed by the same lowercase letter in the columns and uppercase in the rows, do not differ statistically from each other for the Tukey test at the level of 1% probability (p <0.01). The Tukey test was carried out independently for each evaluated period (7 days and 13 days). The averages of the control on the seventh and thirteenth days were the same (5,00 ± 0,00), reaching the maximum growing on the seventh day.
Based on the paired culture method, the mycelial growth (Table 2) was evaluated until 13th day, the isolates 02(F. verticillioides), 03 (Fusarium sp. of the FFSC), 04 (Fusarium sp. of the FFSC), 05 (F. verticillioides), 08 (F. verticillioides) and 09 (F. sulawensiensis) had their growth significantly reduced by the two treatments when compared to the control. Isolates 01 and 10 had growth affected only by T. asperellum, while the growth of the isolates 06 and 07 was not influenced by the treatments. Carvalho et al. (2011) observed similar results, comparing T. harzianum and Fusarium oxysporum f.sp. phaseoli by the pairing culture method, where the averages on the seventh day were lower than control; and on the 13th day, the total colonization of T. harzianum was higher than the pathogen.
Due to the aggressive and faster growth, Trichoderma colonizes primarly the substrate and shows a better efficiency to absorb nutrients than other fungi (Sood et al., 2020) and the competition for carbon and nitrogen, together with competition for position on the substrate, can limit the growth of Fusarium (Vinale et al., 2008). In addition to affect mycelial growth, the antagonist can also impacts the penetration, progression and colonization of pathogen on the substrate, allied with the production of compounds with fungitoxic response (Harman et al., 2004;Boughalleb-M'Hamdi, Salem & M'Hamdi, 2018).
By pairing the culture of T. harzianum and Sclerotium rolfsi, Jana & Mandal (2017) 78.3% of pathogen growth. The efficiency of Trichoderma as an antagonist is due to its rapid development and production of large amounts of conidia, ability to survive in unfavorable environmental conditions; and efficiency in nutrient use. This genus is highly aggressive against pathogens, preventing their development (Benítez et al., 2004;Carreras-villaseñor, Sánchez-Arreguín & Herrera Estrella, 2012). Based on sporulation values (Table 3) of Fusarium isolates paired with Trichoderma, it was observed that the sporulation of 03 (Fusarium sp. of the FFSC), 05 (F. verticillioides) and 07 (Fusarium sp. of the FFSC) isolates was reduced by Trichoderma isolates, whereas isolate 04 (Fusarium sp. of the FFSC) a significant reduction was observed when treated only with T. asperellum. For the other Fusarium isolates, none of the treatments significantly interfered with spore production compared to the control.
The inhibition of sporulation on Fusarium isolates, can be explained by the production of compounds associated with growth inhibition, such as koninginins, viridins and trichoviridins (Reino, Guerrero, Hernández-Galán & Collado, 2008). As sporulation is strategic for the spread of fungi in the environment (Huang & Hull, 2017), reduce it increases the efficiency of management and biological control of plant pathogens. 10.1 ± 4.2 aC 8.8 ± 1.5 aC 9.9 ± 0.8 aC 8 F. verticillioides 17.9 ± 2.2 aA 12.6 ± 1.6 bB 12.5 ± 0.6 bB 9 F. sulawesiensis 17.7 ± 0.2 aA 11.6 ± 1.1 bB 7.0 ± 1.6 cD 10 F. verticillioides 17.5 ± 0.0 aA 14.5 ± 1.1 bA 9.7 ± 0.7 cC Averages followed by the same lowercase letter in the columns and uppercase in the rows, do not differ statistically from each other by the Tukey test (p <0.05).  Zivkovic, Stojanovic, Ivanovic, Gavrilovic & Popovic (2010) observed that spore suspensions of T. harzianum and Gliocladium roseum significantly reduced Colletotrichum acutatum and C. gloesporioides sporulation at 86% and 89%, respectively. The sporulation reduction was observed for some isolates tested in this study. However, the sporulation of six Fusarium isolates was not reduced by the two Trichoderma isolates tested.

Conclusion
Fusarium isolates obtained from fava beans seeds in Paraíba, Brazil, belong to four species -F. sulawesiensis, a possible new phylogenetic species in the F. incarnatumequiseti species complex (FIESC), F. verticillioides and a possible new phylogenetic species in the F. fujikuroi species complex (FFSC). Further studies are necessary to present a proper morphological description and delimitation of these species. Trichoderma sp. and T. asperellum isolates show antagonistic potential against Fusarium isolates. The two Trichoderma isolates tested had a significant impact inhibiting Fusarium isolates growth, but had a significant impact on sporulation of only a few isolates. T. asperellum is indicated as the most promising for the biological control of most Fusarium isolates obtained from fava beans.
Fava beans, cultivated mainly in semiarid areas by family farmers in Brazil, represents an important source of income for the country. In this study, we identified Fusarium species associated with fava beans and indicated ecofriendly alternatives for their control. As we discussed, the treatment with Trichoderma consists of an efficient management strategy, which could enhance the potential of this crop, which is mainly affected by phytopathogens. Further, it is important that studies verify the action of these Trichoderma isolates on the growth of fava bean seedlings.